Key Concepts Learned
- Neighborhood change should be viewed as a dynamic process rather than a single snapshot.
- Neighborhood states can be defined through clustering methods (e.g., k-means) for each time point.
- A sequence for each neighborhood is created by stringing together its state across multiple years.
- Sequence similarity metrics allow comparison of entire change trajectories.
- Hierarchical clustering (such as Ward’s method) groups neighborhoods with similar developmental paths.
- Mapping trajectory clusters reveals spatial patterns of stability, decline, or transformation.
- This approach highlights both the pace and direction of neighborhood change.
Coding Techniques
- Normalizing variables to make them comparable across years.
- Using
kmeans() to assign yearly neighborhood categories.
- Reshaping data to wide format to build sequences.
- Creating sequence objects with tools like
seqdef() from sequence analysis packages.
- Computing distance matrices using sequence similarity measures.
- Applying hierarchical clustering to group neighborhoods into trajectory types.
- Visualizing the spatial distribution of trajectory clusters on maps.
Questions & Challenges
- Choosing the appropriate number of clusters for meaningful interpretation.
- Understanding how parameter choices in sequence distance metrics influence results.
- Translating trajectory groups into clear policy implications rather than descriptive labels.
- Balancing statistical rigor with interpretability for community and policy audiences.
Connections to Policy
- Different trajectory patterns imply distinct policy strategies: stabilization, investment, monitoring, or protection.
- Sequence analysis reveals long-term processes that may be invisible in single-year data.
- Policymakers can identify neighborhoods showing early signs of change or long-term stagnation.
- Understanding trajectories supports more tailored and equitable urban interventions.
Reflection
- This framework shifted my mindset from thinking about neighborhoods as fixed categories to understanding them as evolving stories.
- Looking at sequences helps capture complexity—volatility, stability, and direction—all at once.
- I can apply this approach in future projects by comparing long-term neighborhood patterns and grounding recommendations in multi-year trends.