An explorative analysis of influencing factors on Digitalization in the German county sphere
Explore the docs »
Table of Contents
This is a Github repository for the Master Thesis in Social and Economic Data Science at the University of Konstanz. The digital transformation is one of the greatest challenges Germany is facing, affecting not just urban but also peripheral regions critically. Enhanced public sector digitalization could bring these areas back into focus due to increased pressure in urban centers. But what factors contribute most to the digitalization of the public sector? Various current approaches in research focus on macroeconomic contexts at the state or city level. However comparably little is known about the fundamental interrelationships at the district level and in German rural areas. This work presents an exploratory dataset that utilizes 23 binary items to measure public sector digitalization operationalized by elementary fields of action in German counties. The items are combined into an index using factor scores derived via non-cognitive Item Response Theory. A stratified random sample of German counties is analyzed using a multilevel Bayesian model. In addition to existing literature, a clear correlation between educational drivers and public sector digitalization is observed also on county level. Furthermore, the paper finds that digitalization strategies can be beneficial for the local district public sector digitalization.
When downloading the ZIP File the organisation of the repository should not be changed (e.g. move the shapefiles.) Regardless of the package installations and preparations, r and rstudio as IDE must be installed.
If not already installed on your device or environment, one has to prepare some basic steps. Rtools is needed in order to work with tools from many developers. See rtools for a guide on installation. Another package, that is important for working more efficiently with packages in r is devtools.
- rtools
- Devtools example
install.packages("devtools") - Brms and brmstools installation
devtools::install_github("paul-buerkner/brms") devtools::install_github("mvuorre/brmstools")
- Troubleshooting: I encountered some trouble when installing some packages or updating the packages using pacman. Manually installing/loading the packages sometimes did not help. A quick fix for me was to remove and re-install the rlang package or manually install the matrix package.
There are four different sources of data in my project
- Inkar_all_counties -> represents all the counties except kreisfreie städte together with relevant contextual data
- strata_sample -> the stratified randomly sampled counties populated with the 23 different binary items
- Shape.zip -> needs to be unzipped first, but not moved. Incorporates the shapefiles for geometric data on German counties and federal states
- Broadband -> Percentage of local population with access to at least 50 Mbit on the German county level
Felix Caspari - felix.caspari@uni-konstanz.de