TY - JOUR
T1 - Multivariate analysis of socioeconomic profiles in the Ruhr Area, Germany
AU - Lengyel, Janka
AU - Roux, Stéphane
AU - Alvanides, Seraphim
N1 - Funding information:
J.L. acknowledges funding from the Mercator foundation within the NEMO Project and would like to thank both the project's team and its strategic partners; the Regional Planning Association Ruhr (RVR) and the Emschergenossenschaft (EGLV) for providing extensive data as well as continuous help and feedback.
PY - 2022/12/1
Y1 - 2022/12/1
N2 - The aim of this article and associate map is to highlight the social and economic diversity of the Ruhr area in Germany through the use of multivariate analysis and visualisation. To this end we combine two different datasets. Demographic parameters stemming from the 2011 German census and socioeconomic indicators obtained from the microdialog of the German post service. Due to the different spatial resolution of the two datasets, we aggregated the data at the neighbourhood (Stadtteil) level. The multivariate analysis was carried out at this scale using Self-Organizing Maps (SOM), an artificial neuron network, which uses an unsupervised learning mechanism for projecting multidimensional data in a low (in our case two) dimensional space. First we used a visualization technique to better comprehend the relationship between our observations via reducing the dimensionality or complexity of our input data. At the same time, we established a global statistical relationships between the indicators. Finally, using these results we built clusters for revealing the distribution of socioeconomic profiles over the whole region. Our results demonstrate that structural inequalities resulting from the processes of industrialization/deindustrialization in the region are still widely persistent and result in characteristic patterns along the three main rivers, the Lippe, Emscher and the Ruhr. In close connection with this, three types of societal segregation patterns become clearly evident in the Ruhr area, namely nationality, age and economic power.
AB - The aim of this article and associate map is to highlight the social and economic diversity of the Ruhr area in Germany through the use of multivariate analysis and visualisation. To this end we combine two different datasets. Demographic parameters stemming from the 2011 German census and socioeconomic indicators obtained from the microdialog of the German post service. Due to the different spatial resolution of the two datasets, we aggregated the data at the neighbourhood (Stadtteil) level. The multivariate analysis was carried out at this scale using Self-Organizing Maps (SOM), an artificial neuron network, which uses an unsupervised learning mechanism for projecting multidimensional data in a low (in our case two) dimensional space. First we used a visualization technique to better comprehend the relationship between our observations via reducing the dimensionality or complexity of our input data. At the same time, we established a global statistical relationships between the indicators. Finally, using these results we built clusters for revealing the distribution of socioeconomic profiles over the whole region. Our results demonstrate that structural inequalities resulting from the processes of industrialization/deindustrialization in the region are still widely persistent and result in characteristic patterns along the three main rivers, the Lippe, Emscher and the Ruhr. In close connection with this, three types of societal segregation patterns become clearly evident in the Ruhr area, namely nationality, age and economic power.
KW - Multivariate analysis
KW - segregation patterns
KW - self-organizing maps
KW - socioeconomic clusters
UR - http://www.scopus.com/inward/record.url?scp=85135769556&partnerID=8YFLogxK
U2 - 10.1080/17445647.2022.2098839
DO - 10.1080/17445647.2022.2098839
M3 - Article
SN - 1744-5647
VL - 18
SP - 576
EP - 584
JO - Journal of Maps
JF - Journal of Maps
IS - 3
ER -