Creation of the 2021/2 Output Area Classification
Metadata.R- creating reference tables and glossaries for Census data.Downloading_data.R– downloading 2011 and 2021 Census data from NOMIS usingnomisrpackage.Comparing_Censuses.R- data cleaning and amalgamation.Transforming_Census_data.R– data manipulation and transformation for the classification.NI.R– modelling 2021 data for Northern Ireland.Correlation.R- testing correlation between the variables.Pre-clustering.R- preparing data for clustering.Clustering.R- clustering of the data.Post-clustering.R- creating maps and plots of the cluster solution.Testing_clustering.RClustergrams.ipynb– creating Clustergrams in Python. (Credits: Prof Alex Singleton)Industry.R- loading Industry dataIndustry_classification.R- creating geodemographic classification with Industry variablesGraph_comparisons.R- comparing data with graphs
List of folders (subfolders & files) in the project:
- API - Census data downloaded and saved with use of
nomisrpackage. - Clean – amalgamated data ready for the analysis. Downloadable via this link.
- Raw_counts - datasets with raw counts
- Percentages - datasets transformed into percentages
- Transformed - datasets transformed with IHS (analysis-ready)
- Final_variables - datasets with OAC variables only
- All_data_clustering - results of the clustering for all investigated datasets.
- Clustering - datasets with cluster assignment for the UK and centroids.
- Lookups - reference tables for 2011 and 2021 Census variables.
- NISRA 2021 - 2021 Census data at LGD level for Northern Ireland.
- Objects - R objects created and stored to ensure consistency of the results or load big files
- SIR – contingency tables on disability counts by age, utilised for calculation of Standardised Illness Ratio.
Downloadable via this link.
- Bar_plots - Comparison of clusters to the UK (as well as Supergroup and Group averages)
- Clustergrams – plots used to establish number of clusters at each classification level.
Downloadable via this link.