Interaction of Spatial and Social Structure

Anna Brazdova, Martin Fleischmann, Lisa Winkler

Charles University

About me

2nd year Ph.D student at the department of Social geography and Regional Development

Faculty of Science at Charles University in Prague

Interplay between population and built form

Understand built-up patterns of housing types and the ability of different population to inhabit them.

Relationship between social structure and built form

Spatial configuration, urban design, and morphology actively shape social interactions, mobility, and segregation

Qualitative methods and quantitative methods

Urban morphometrics - deprivation prediction

Rarely been empirically measured

How strong are the associations between built form and population

Systematic exploration of the associations

How to measure the relationship between social structure and the built form?

Does the relationship differ across different built form types?

Is the relationship consistent or spatially variable?

Case Study of Czechia

residential differentiation

driving forces

Source

Built form types as the dependent variable

Morphometric classification

Quantitative identification of built form types.

Based on similar morphological characteristics shared by street segments and building footprints.

Focuses on geometry and spatial configuration within the built fabric.

Level 3 of the classification - 7 main housing types in Czechia

Coherent Interconnected Fabric

Source

Coherent Dense Adjacent Fabric

Source

Coherent Dense Disjoint Fabric

Source

Incoherent Large-scale Homogeneous Fabric

Source

Incoherent Small-scale Linear Fabric

Source

Incoherent Small-scale Sparse Fabric

Source

Incoherent Small-scale Compact Fabric

Source

Scale

Census characteristics as the independent variables

Census variables

Population density

Age structure

Employment sector

Employment status

Dwelling ownership

Education level

Household size

Citizenship

Religion

Marital status

Source

Census processing

Global modelling

Logistic Regression - 0.098

Random Forest Classification - 0.377

Built form types

Prediction

Error

Spatial autocorellation of the error

Geographically weighted modelling

Global models do not account for the geographic variation in the relationship.

Geographically weighted models capture this by applying local models rather than a single global model.

Geographically weighted classification

Similar in concept to Geographically Weighted Regression (GWR).

Categorical or class-based outcomes.

Separate classifier for each location using data weighted by geographic proximity.

Weighting

Controlled by a distance-decay parameter.

Nearby observations are given more weight than distant ones.

Illustration of bandwidth and its relation to weight, Fotheringham et al. (2002, 44–45)

Bandwidth

Controls the spatial scale over which a process varies.

Conceptual diagram explaining fixed (left) and adaptive weighting (right) schemes. Sachdeva, M., & Fotheringham, A. S. (2020)

Binary Classification

The distribution of built form classes is uneven across space.

Some built forms do not appear in certain locations at all.

Each model can be tuned to local prevalence and have custom thresholds, weights, bandwidth…

Results

Results

Results

Incoherent Small-Scale Sparse Fabric

Incoherent Small-Scale Sparse Fabric

Incoherent Small-Scale Sparse Fabric

Conclusion

Locally linear but globally heterogeneous.

Specific relationship between social structure and different built form types.

In some locations, education, ownership regime and employment status strongly increase the probability of belonging to specific built forms.

Preliminary results, would appreciate some feedback!

Thank you

anna.brazdova@natur.cuni.cz

www.linkedin.com/in/kryndlea