applied-regression-activity

Coursework from Criminological Research Methods Workshops, Lent Term, MPhil Criminology at the University of Cambridge, taught by Professor Charles Lanfear.

Data Description

In this workshop, I created an LSOA-level dataset for London. I merged London Crime Data (Source: data.police.uk, 2022), LSOA Indices of Deprivation (Source: opendatacommunities.org, 2019), and LSOA Population Density (Source: data.london.gov.uk, 2014).

A Lower-layer Super Output Area (LSOA) is a small geographic unit used for statistical analysis in England and Wales ¹.

This project involves data construction, cleaning, visualization, modeling, model comparison, and interpretation. I assess how deprivation is associated with violence, both independently and controlling for population density.

File Information

• data: (folder)
  • Analysis: final analysis used for modeling and visualization
  • Derived: cleaned and constructed datasets (crime data, population density, deprivation)
  • Raw: original datasets (London Crime Data, October-December 2022)
• applied-regression-activity.R: Includes code for constructing the LSOA dataset by cleaning, transforming, and merging crime, deprivation, and population density data
• applied-regression-activity.Rproj: RStudio project file
• applied-regression-visualization-and-modeling.R: Includes code for constructing visualizations, linear regression, model comparison, and interpretation.
• imd2019lsoa.csv: LSOA Indices of Deprivation 2019 dataset
• land-area-population-density-lsoa11-msoa11.xlsx: Population density dataset

Other

• .gitignore
• README.md

Footnotes

1. https://ocsi.uk/term/lower-layer-super-output-area-lsoa/ ↩