Information

This repo is for Python scripts for "Capturing urban diversity through languages: long-term changes in multilingual residential neighbourhoods in the Helsinki Metropolitan Area" published in Population, Space and Place.

Pre-analysis steps

• Apply for access to individual-level register data of Statistics Finland
• Calculate linguistic diversity of register data per grid cell with scikit-bio similar to how it was done in a previous article or with language family information in this article's appendices with calculate_langfam_diversity.py.

Suggested running order of scripts

Step	Script	Description	Input	Output
1	calculate_langfam_diversity.py	Calculates diversity of language families per grid cell	Output from Step 1	Geopackages
2	add_langfam_genus.py	Adds information on the family and genus per each language	First language information per individual in a CSV file	Language family and genus information per individual
3	calculate_speakers.py	Calculates the number of speakers per a grid cell across the language groups used in the article	First language information per individual in a CSV file	Geopackage
4	calculate_language_changes.py	Calculates how many times individuals have changed their language	First language information per individual in a CSV file	Pickled dataframe on changes
5	HMA_overall_diversityplot.py	Plots subplots in Figure 1 and full Figure 2	Individual-level data from Statistics Finland	PDF Figures
6	get_gridcell_histories.py	Gets annual linguistic diversity values per grid cell	Geopackage of diversity information per grid cell	Geopackage of metric-specific grid cell histories
7	calculate_livingspace_commutes.py	Aggregates the records on living space and commute distances to grid cells	FOLK commuting and employment statistics	Pickled data frame
8	add_livingspace_commute_to_HMAgrid.py	Adds the calculated values from Step 5 to the spatial grid data	Step 5	Geopackage
9	plot_diversities_in_new_old_stable_grids.py	Provides non-normalized plots for Figure 5	Output from step 4	PDF Graph and pickled dataframes
10	norm_grid_trajectories.py	Calculates the normalized values of grid trajectories	Geopackage of diversity information per grid cell	PDF Graphs and pickled dataframes
11	popweigh_grid_trajectories.py	Calculates the population-weighted average values of grid trajectories	Geopackage of diversity information per grid cell	PDF Graphs and pickled dataframes
12	plot_grid_diversitiyes_figure5.py	Plots final Figure 5	Outputs from steps 8 and 9	PDF Graph
13	plot_divs_est_som_87-19.py	Plots figures 9 and 10, yields outputs	Output from Step 1	Figures 9 and 10, dataframes
14	plot_weighted_divs_est_som_87-19.py	Plots Figure 4	Output from Step 2	Geopackage with stability classficiations
15	calculate_spatial_markov.py	Calculate and plot Markov chain matrices	Output from Step 5	PDF matrices (Figure 6) and matrices as pickled dataframes
16	compare_markov.py	Compares Markov probability matrices with Jensen-Shannon distances	Outputs Figure 7	PNG file

Other information

Some of these scripts have not been formatted to be flexibly used on all computers. You might have to alter some hard-coded filepaths.